Using AI to predict the risk of lightning caused fires with Dmitri Kalashnikov | EP 26 | 3reate

Dmitri Kalashnikov is a climate scientist and National Science Foundation postdoctoral research fellow at the University of California-Merced. Having created Convolutional Neural Network-based predictive models for lightning in the western United States; he is currently working on applying these predictive models to climate models to project future changes in the climatology of lightning storms in the western United States.

Ever wonder how climate science actually impacts your life, especially with wildfire season becoming a new normal? 🌲🔥 Andrew reconnects with postdoctoral climate scientist Dmitri Kalashnikov, who pulls back the curtain on his vital research. They dive into how political policies and funding cuts threaten the scientific foundation we all rely on. Dmitri breaks down the rigorous, multi-year process of publishing a scientific paper and explains his groundbreaking new research that uses AI to predict future risk of lightning-caused fires. From the “Ostrich Effect” in politics to the real-world value of predicting extreme weather, this conversation reveals why the work of scientists is more critical than ever.

Read what Dmitri is reading: Doppelgänger by Naomi Klein

Read Dmitri’s paper Projections of Lightning-Ignited Wildfire Risk in the Western United States here.

Read Dmitri’s interview with the Guardian: Climate crisis will increase frequency of lightning-sparked wildfires, study finds

Read Dmitri’s interview with the New York Times here: Lightning Strikes California With Unusual Frequency This Summer

Watch what Andrew is watching: Ron Deibert from The Citizen Lab speaks about his new book Chasing Shadows: Cyber Espionage, Subversion, and the Global Fight for Democracy

Support the pod: https://3reate.com, https://ko-fi.com/3reate or https://patreon.com/3reate

Listen to Dmitri’s pervious 3reate Podcast, Episode 13

Chapters:

(0:00) Welcome Back, Dmitri: Andrew welcomes postdoctoral climate scientist Dmitri back to the podcast to discuss the state of climate science in 2025.

(1:41) Government’s Real Impact on Science: Dmitri explains that actual government funding cuts are a bigger problem for scientists than what the media says.

(4:31) The Hidden Value of Public Research: The hosts discuss how foundational government-funded science, from the internet to AI, creates massive, long-term economic value.

(6:33) How Climate Science Affects You: Dmitri breaks down how his research on wildfire risk directly informs public awareness and shapes critical policy decisions.

(14:19) What Is a Scientific Paper?: Get a behind-the-scenes look at the long and rigorous process of writing, peer-reviewing, and publishing an academic paper.

(26:01) Predicting Wildfires with AI: Dmitri details his groundbreaking new research that uses AI and neural networks to project the future risk of lightning-caused fires.

(35:21) The Data Behind the Models: Learn about “atmospheric reanalysis,” a fascinating process where NASA combines data from countless sources to create a complete weather history of the planet.

(49:40) When Science News Goes Wrong: Dmitri uses the “wavy jet stream” hypothesis to illustrate how the media can sometimes run with exciting but unproven scientific theories.

(1:01:43) The Future of Climate Research: Dmitri gives a sneak peek into his next projects, including the connection between heat waves and wildfire intensity.

(1:06:34) Book Recommendation & Final Thoughts: Dmitri recommends Naomi Klein’s book “Doppelganger” for its insights into social media and political polarization.

3reate
3reate
Using AI to predict the risk of lightning caused fires with Dmitri Kalashnikov | EP 26 | 3reate
Loading
/